Organochlorines such as DDT, its metabolite DDE, and polychlorinated biphenyls (PCBs) are persistent organic pollutants that may be toxic at levels encountered by the general population, through exposure in the food chain. Each organochlorine has one or more types of biologic activity, and corresponding potential mechanisms of action. Most of my work in this area is based on analysis of serum from pregnant women who were in a cohort study named the Collaborative Perinatal Project (CPP). Here I describe some of my results from the CPP study. Results from other studies I have done on the health effects of persistent organic pollutants are indicated in the list of publications. Sera from over 50,000 pregnant women in the CPP have been frozen at -20 degrees Centigrade since 1959-1965, and with the health data on their offspring constitute a resource that is still actively used by many investigators. We selected a subset of about 2,800 children born in 1959-1966 and determined organochlorine concentrations in the corresponding stored maternal pregnancy sera, in 1997-1999. In earlier CPP papers we examined DDE levels in relation to fetal loss in previous pregnancies, and, in the CPP pregnancy, preterm birth, small-for-gestational-age, birthweight, and subsequent height; in all cases we found associations suggestive of adverse effects. To look for evidence of androgen-blocking, we examined maternal DDE level during pregnancy in relation to adjusted odds of cryptorchidism and hypospadias among their male offspring; for these outcomes the results were equivocal. Furthermore, we evaluated prenatal PCB exposure in relation to developmental assessment performance at 8 months of age, 7 years of age, and hearing as reflected by audiometric examination at age 8 years. We also examined exposure to DDT and DDE in relation to IQ at age 7 years. None of these studies showed any associations. Last year's progress We published a study on the relation of organochlorines to preeclampsia (no association). We used data from the CPP as an example in a study of statistical methods that will be useful in studies of mixtures of organochlorines (published). We examined how much of the PCB-birthweight association that has been reported in meta-analyses is attributable to pharmacokinetic confounding by pregnancy weight gain (published). We also used expertise on the pharmacokinetics of organochlorines to assist the Environmental Protection Agency on approaches to risk assessment (published).